DD23 Table Profile: how it was built
Can I trust this table, and which of its columns can I use? Rebuild this dashboard step by step in the Data Canvas.
DD23 · Table Profile
Code in this build: 6 widgets written in code (1 in SQL, 5 in Groovy), and 2 filter, chart or pivot settings written as code.
The question: Can I trust this table, and which of its columns can I use?
Who asks, and when: The analyst meeting a new table.
The decision: Whether the table can be trusted, and which columns can be used.
What the dashboard shows: Pick any table and you know it in half a minute: support_tickets holds 14,000 rows over 15 columns, 476 of them opened in the last 30 days, and 4 columns have gaps - order_id is empty on 52.64% of tickets, because a ticket need not be about an order, and csat_score on 58.33%, because most people never answer the survey. Its priority column is 51.56% normal and 8.21% urgent. Point the same page at orders.total_amount and the chart turns into twenty ranges over 23,162 distinct amounts, 20,782 orders inside the first one.
Follow the pictures below, in order, and you end up with the same dashboard. Every step is one thing you do in the Data Canvas, on the demo data that installs with DataPallas.